https://github.com/nixtla/mlforecast - v1.0.2

Bug Fixes

• fix(compat): handle zero offset in shift_array @jmoralez (#480)
• fix(global-sklearn-tfm): apply inverse transform to each column @jmoralez (#477)

- Python
Published by github-actions[bot] about 1 year ago

https://github.com/nixtla/mlforecast - v1.0.1

Bug Fixes

• fix: X_df handling in direct approach @jmoralez (#468)
• fix(auto): remove invalid params from xgboost default space @jmoralez (#464)

- Python
Published by github-actions[bot] about 1 year ago

https://github.com/nixtla/mlforecast - v1.0.0

Breaking Change

• breaking: remove window_ops and numba dependencies @jmoralez (#462)

- Python
Published by github-actions[bot] about 1 year ago

https://github.com/nixtla/mlforecast - v0.15.1

Changes

• chore: deprecate window_ops @jmoralez (#410)

New Features

• feat(distributed): support ids in predict @jmoralez (#454)
• feat(auto): support input_size @jmoralez (#451)

- Python
Published by github-actions[bot] about 1 year ago

https://github.com/nixtla/mlforecast - v0.15.0

Breaking Change

• breaking: drop rows with null targets when dropna=False @jmoralez (#447)

Bug Fixes

• fix(auto): support custom column names @jmoralez (#449)

Enhancement

• enh(distributed): propagate null features in spark @jmoralez (#448)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.14.0

New Features

• feat: add weight_col to MLForecast.fit and MLForecast.cross_validation @jmoralez (#444)
• feat: infer samples required for built-in lag transforms updates @jmoralez (#445)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.6

Bug Fixes

• fix(distributed): exogenous handling in distributed cross validation @jmoralez (#443)
• fix(distributed): support pre-computed features @jmoralez (#436)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.5

Enhancement

• enh: add step_size to AutoMLForecast @jmoralez (#426)
• support step_size selection in optimization.mlforecast_objective @bchaoss (#419)
• use TypeVar for dataframes and distribute py.typed file @jmoralez (#408)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.4

New Features

• feat: mlflow flavor @jmoralez (#406)

Documentation

• Clear up the README for the new user @Ammar-Azman (#397)

Enhancement

• make season_length optional in AutoMLForecast @jmoralez (#399)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.3

Bug Fixes

• handle no target transforms in DistributedMLForecast.to_local @jmoralez (#388)

Enhancement

• ensure static features are constant @jmoralez (#391)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.2

New Features

• support prediction intervals in auto @jmoralez (#370)

Bug Fixes

• remove dots from feature names in distributed @jmoralez (#382)
• fix minsamplessplit in random forest space @jmoralez (#380)

Enhancement

• store prediction intervals inputs in MLForecast.save @jmoralez (#383)
• support polars in GlobalSklearnTransformer @jmoralez (#377)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.1

Dependencies

• add polars extra @jmoralez (#368)
• support polars 1.0 @jmoralez (#366)

Enhancement

• add fitted argument to AutoMLForecast.fit @jmoralez (#351)

- Python
Published by github-actions[bot] over 1 year ago

https://github.com/nixtla/mlforecast - v0.13.0

Breaking Change

• set refit=False and results_ as dict in AutoMLForecast @jmoralez (#341)

Bug fixes

• fix: fitted nonrecursive cv with horizon >= 10 @adriaanvh1 (#333)

Enhancement

• speedup date features @jmoralez (#340)
• Create CODEOFCONDUCT.md @tracykteal (#335)

- Python
Published by github-actions[bot] almost 2 years ago

https://github.com/nixtla/mlforecast - v0.12.1

New Features

• add auto module for hyperparameter optimization @tblume1992 (#306)
• add DistributedMLForecast.update @jmoralez (#324)

Bug Fixes

• fix cv fitted values with prediction intervals @jmoralez (#330)

- Python
Published by github-actions[bot] almost 2 years ago

https://github.com/nixtla/mlforecast - v0.12.0

Enhancement

• migrate to coreforecast @jmoralez (#311)

- Python
Published by github-actions[bot] almost 2 years ago

https://github.com/nixtla/mlforecast - v0.11.8

Bug Fixes

• ensure coreforecast is installed for AutoDifferences @jmoralez (#314)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.7

New Features

• add auto differences @jmoralez (#310)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.6

New Features

• add to_local method to distributed forecast @jmoralez (#302)
• support saving and loading forecast objects @jmoralez (#301)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.5

Bug Fixes

• add update method to target_transforms @jmoralez (#293)

Enhancement

• use coreforecast target_transforms when installed @jmoralez (#294)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.4

Bug Fixes

• fix predict with multiple models @jmoralez (#290)

Dependencies

• polars updates @jmoralez (#291)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.3

New Features

• add level argument to forecastfittedvalues @jmoralez (#287)
• add X_df argument to distributed predict @jmoralez (#286)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.2

New Features

• add X_df debugging methods @jmoralez (#283)
• add quantile lag_transforms @jmoralez (#282)
• support lag transforms namer @jmoralez (#280)

Documentation

• add sklearn pipelines guide @jmoralez (#277)

Dependencies

• update utilsforecast @jmoralez (#281)

Enhancement

• don't recompute features already present in df @jmoralez (#279)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.1

Bug Fixes

• fix RollingStd typo @jmoralez (#276)

Enhancement

• Improve error message "Found missing inputs in X_df" @MarcoGorelli (#273)
• use backtest_splits from utilsforecast @jmoralez (#271)

- Python
Published by github-actions[bot] about 2 years ago

https://github.com/nixtla/mlforecast - v0.11.0

New Features

• support lag transformations from coreforecast @jmoralez (#265)
• add feature_engineering module @jmoralez (#261)
• add as_numpy argument @jmoralez (#249)
• add polars support @jmoralez (#241)

Breaking Change

• remove deprecated arguments @jmoralez (#240)
• remove dynamic_dfs argument @jmoralez (#239)

Bug Fixes

• deterministic column order @jmoralez (#262)
• fix inverse transforms for fitted values when series were dropped @jmoralez (#255)
• Fix distributed cv @jmoralez (#254)
• add packaging to dependencies @jmoralez (#235)

Documentation

• add as_numpy guide @jmoralez (#258)
• add analyzing models and custom training how-to guides @jmoralez (#236)

Enhancement

• keep df order in cv @jmoralez (#257)
• handle short series exception @jmoralez (#256)
• support polars dataframe in TimeSeries.update @jmoralez (#252)
• issue warning instead of error for short series in cv @jmoralez (#247)
• ensure lags are positive integers @jmoralez (#232)

Full Changelog: https://github.com/Nixtla/mlforecast/compare/v0.10.0...v0.11.0

- Python
Published by github-actions[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.10.0

Breaking Change

• remove differences argument @jmoralez (#215)

Bug Fixes

• fix X_df slices @jmoralez (#228)

Documentation

• move distributed API reference to quickstart @jmoralez (#229)
• extract how-to guides from API reference @jmoralez (#224)
• Feature/electricity load forecasting (PJM) tutorial using MLForecast @uumami (#208)
• Prediction intervals for machine learning models @Naren8520 (#196)

- Python
Published by github-actions[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.9.3

Bug Fixes

• support fitted inverse transform in LocalStandardScaler @jmoralez (#206)
• fix fitted_values methods @jmoralez (#198)

Enhancement

• raise error when maxhorizon and models don't match @jmoralez (#204)

- Python
Published by github-actions[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.9.2

New Features

• add forecastfittedvalues method @jmoralez (#190)
• support integer refit in cross_validation @jmoralez (#189)
• add GlobalSklearnTransformer @jmoralez (#187)

- Python
Published by release-drafter[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.9.1

New Features

• support predicting a subset of series @jmoralez (#183)

Enhancement

• raise informative error when interpreting dynamic features as static @jmoralez (#182)

- Python
Published by release-drafter[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.9.0

Enhancement

• faster MLForecast.preprocess @jmoralez (#179)
• deprecate dynamicdfs argument in favor of Xdf @jmoralez (#176)
• improve staticfeatures definition @jmoralez (#175)

- Python
Published by release-drafter[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.8.1

Bug Fixes

• fix TimeSeries.update method @jmoralez (#173)
• fix static_features order @jmoralez (#174)

- Python
Published by release-drafter[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.8.0

New Features

• Add LocalStandardScaler @jmoralez (#171)

Enhancement

• make argument names compatible with other nixtla libraries @jmoralez (#166)

Bug Fixes

• fix keeplastn with target_transforms @jmoralez (#171)

- Python
Published by release-drafter[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.7.4

New Features

• add cross validation fitted values @jmoralez (#164)
• allow idcol in staticfeatures @jmoralez (#161)

Enhancement

• raise error for wrong frequency in cross_validation @jmoralez (#160)
• Add new release drafter @FedericoGarza (#146)
• Add new issue template @FedericoGarza (#145)

- Python
Published by release-drafter[bot] over 2 years ago

https://github.com/nixtla/mlforecast - v0.7.3

Enhancement

• extend custom metric signature in LightGBMCV and add example @jmoralez (#137)

- Python
Published by release-drafter[bot] almost 3 years ago

https://github.com/nixtla/mlforecast - v0.7.2

Bug Fixes

• use target_col in conformity scores @jmoralez (#134)

Enhancement

• Allow intervals for horizon lower than window size @FedericoGarza (#129)

- Python
Published by release-drafter[bot] almost 3 years ago

https://github.com/nixtla/mlforecast - v0.7.1

New Features

• add TimeSeries.update method to update target values @jmoralez (#119)

Documentation

• fix slack link in README @mergenthaler (#117)

Maintenance

• set lower bound on spark for tests @jmoralez (#118)

Enhancement

• remove dynamic_dfs argument from LightGBMCV when it can be inferred @jmoralez (#125)

- Python
Published by release-drafter[bot] almost 3 years ago

https://github.com/nixtla/mlforecast - v0.7.0

New Features

• add target_transforms @jmoralez (#110)
• add ray integration @FedericoGarza (#104)
• add inputsize argument to crossvalidation @jmoralez (#107)
• add fugue backend for distributed training with spark and dask @jmoralez (#90)
• add conformal distribution strategy @FedericoGarza (#97)

Breaking

• remove id_col='index' and set defaults for column names @jmoralez (#114)
• remove Forecast object @jmoralez (#113)
• replace dask-based distributed forecast with fugue-based @jmoralez (#102)

Documentation

• improve readme @FedericoGarza (#111)
• add fugue to docs @jmoralez (#100)
• add transfer learning tutorial @FedericoGarza (#93)
• fix prediction intervals plot @FedericoGarza (#92)
• Add prediction intervals tutorial @FedericoGarza (#87)

Maintenance

• set encoding on README open @jmoralez (#112)
• split distributed tests in CI @jmoralez (#99)

Enhancement

• extract distributed fit logic to model classes @jmoralez (#103)
• vectorize prediction intervals creation @jmoralez (#101)

- Python
Published by release-drafter[bot] almost 3 years ago

https://github.com/nixtla/mlforecast - v0.6.0

New Features

• Add prediction (conformal) intervals @FedericoGarza (#86)
• Add nbdev merge to gitattributes @FedericoGarza (#85)

Bug Fixes

• remove lightgbm import from project namespace @jmoralez (#88)

Maintenance

• automate release @jmoralez (#89)

- Python
Published by release-drafter[bot] about 3 years ago

https://github.com/nixtla/mlforecast - v0.5.0

Breaking changes

• remove dashes from feature names by @jmoralez in https://github.com/Nixtla/mlforecast/pull/69
• replace predict_fn with callbacks by @jmoralez in https://github.com/Nixtla/mlforecast/pull/73

Features

• add MLForecast.from_cv by @jmoralez in https://github.com/Nixtla/mlforecast/pull/71
• allow models to be dict by @jmoralez in https://github.com/Nixtla/mlforecast/pull/72
• Add step size argument to cross validation method by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/74
• Add new_data argument to predict method (allow transferability) by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/79
• Perform cross validation without refitting the models by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/81
• Support one model per horizon approach by @jmoralez in https://github.com/Nixtla/mlforecast/pull/80
• support multiple models in cross_validation by @jmoralez in https://github.com/Nixtla/mlforecast/pull/84 ## Bug fixes
• Remove dynamic_dfs argument from cross_validation method by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/82 ## Documentation
• add getting started docs section by @jmoralez in https://github.com/Nixtla/mlforecast/pull/64
• Add cross-validation tutorial by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/76
• Add electricity peak forecasting tutorial by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/77
• Improve description preprocessing ERCOT dataset by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/78 ## Maintenance
• set methods on GroupedArray and preserve idcol in TimeSeries.fittransform by @jmoralez in https://github.com/Nixtla/mlforecast/pull/70

- Python
Published by jmoralez about 3 years ago

https://github.com/nixtla/mlforecast - v0.4.0

What's Changed

• rename Forecast to MLForecast by @jmoralez in https://github.com/Nixtla/mlforecast/pull/63

Full Changelog: https://github.com/Nixtla/mlforecast/compare/v0.3.1...v0.4.0

- Python
Published by jmoralez about 3 years ago

https://github.com/nixtla/mlforecast - v0.3.1

What's Changed

• fix unused arguments by @jmoralez in https://github.com/Nixtla/mlforecast/pull/61

Full Changelog: https://github.com/Nixtla/mlforecast/compare/v0.3.0...v0.3.1

- Python
Published by jmoralez over 3 years ago

https://github.com/nixtla/mlforecast - v0.3.0

What's Changed

• raise error when serie is too short for backtest by @jmoralez in https://github.com/Nixtla/mlforecast/pull/32
• allow models list by @jmoralez (#34, #36)
• [FEAT] Allow used by GitHub section hardcoding lib name by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/37
• [FIX] Add black as a development dependency by @FedericoGarza in https://github.com/Nixtla/mlforecast/pull/38
• rename backtest to cross_validation and return single dataframe by @jmoralez in https://github.com/Nixtla/mlforecast/pull/41
• Remove TimeSeries from Forecast constructor by @jmoralez in https://github.com/Nixtla/mlforecast/pull/44
• allow passing column names as arguments. allow ds to be int by @jmoralez in https://github.com/Nixtla/mlforecast/pull/45
• add LightGBMCV by @jmoralez in https://github.com/Nixtla/mlforecast/pull/48
• support applying differences to series by @jmoralez in https://github.com/Nixtla/mlforecast/pull/52
• allow functions as date features by @jmoralez in https://github.com/Nixtla/mlforecast/pull/57
• Improve docs by @jmoralez in https://github.com/Nixtla/mlforecast/pull/59

New Contributors

• @FedericoGarza made their first contribution in https://github.com/Nixtla/mlforecast/pull/37

Full Changelog: https://github.com/Nixtla/mlforecast/compare/v0.2.0...v0.3.0

- Python
Published by jmoralez over 3 years ago